Logical analysis of data for estimating passenger show rates at Air Canada

This paper examines an approach for improving the accuracy of passenger show rate predictions for airlines. The method involves the logical analysis of data. This method based on optimization techniques, detects sets of conditions for which all the satisfying passengers have a significantly higher or lower show rate than the studied population. The passengers are classified according to the patterns as show, no-show or unknown. When compared to Air Canada’s current tool for overbooking forecasts, which is based on historical statistics, logical analysis of data model appears to be very competitive.